CVS difference for ais/ai-00295.txt

Differences between 1.1 and version 1.2
Log of other versions for file ais/ai-00295.txt

--- ais/ai-00295.txt	2002/05/25 03:42:19	1.1
+++ ais/ai-00295.txt	2002/06/11 05:15:50	1.2
@@ -728,3 +728,38 @@
+From: Robert Dewar
+Sent: Saturday, May 25, 2002  8:28 AM
+<<Of course, but it is relatively easy to disprove the assertion that no one
+uses access subtypes on general access types. Simply provide a significant
+example. Proving the assertion is much harder, of course, but I'd like to
+know whether we need to put some effort in here.>>
+That's the wrong point of view. If current compilers are allowing this,
+then programs may or may not be using significant or insignificant
+examples of this usage. I know this is FUD, but we have to be very
+sure that it is worth while making something illegal that was previously
+legal, since it can be a huge pain.
+From: Randy Brukardt
+Sent: Friday, June 7, 2002  6:54 PM
+No, that is the correct point of view, because I am already convinced that it
+is worthwhile to make this change. (The alternatives being much worse, IMHO,
+and it fixes other problems at the same time.) That may not be true for others.
+The main question in my mind is whether it is too incompatible -- that is, does
+it break too many existing programs.
+If no one can disprove the assertion, then I would probably spend the "huge"
+(as you put it) effort to try to find out, by creating a tool that can be run
+on large volumes of code by users to determine the usage of this (and probably
+other) features, and then distributing it as widely as possible. However, if
+there are significant uses of the feature such that there would be a lot of
+opposition to the change, I'd rather spend the time on something productive
+(like my resume :-).

Questions? Ask the ACAA Technical Agent